Story perspectives

Anthropic and OpenAI Propose AI Safety, Urge Slower Development

By Drooid · Sep 18, 2026, 2:41 PM UTC · How we work

Summary

Story summary

Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman proposed safety initiatives at Dreamforce 2026, urging slower AI development.
Nvidia CEO Jensen Huang urged labs to run as fast as they can at Dreamforce.
Salesforce’s Agentforce tools use last-year AI models instead of frontier models like Claude Fable 5.1.
Attendees noted older, cheaper AI models remain sufficient for most sales and customer-service applications.

1 of 1

Story summary

Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman proposed safety initiatives at Dreamforce 2026, urging slower AI development.
Nvidia CEO Jensen Huang urged labs to run as fast as they can at Dreamforce.
Salesforce’s Agentforce tools use last-year AI models instead of frontier models like Claude Fable 5.1.
Attendees noted older, cheaper AI models remain sufficient for most sales and customer-service applications.